Am 01.11.2004 um 08:56 schrieb Blake:
On Mon, 1 Nov 2004 08:33:37 +0100, Marcus Denker denker@iam.unibe.ch wrote:
There is *nothing* going on at that times on the list wrt. harvesting. Just normal talking on IRC.
Well, I didn't mean that I'd picked it up from listening on IRC. I don't do IRC. I meant I'd picked up on it here.
Harvesting would mean: Lets fire up BFAV. Look at a submission, test, look the code, decide if that should go in.
Sounds simple. On what basis does one make that decision?
That's the hard part... I don't think it makes sense to have a hard definition for that. We tried that, and it turned into a burocratic nightmare.
For all submissions: Is there an attachement? are there conflicts with updates? Does the .cs has a preamble?
For fixes: Does it fix the bug? Who did the fix? does it look ok? Tests? For enh: Does it make sense to have it in the image or should it be on SM? Who did it? does the code look ok? commented? tests? does it work? Do i want to see this in the image?
stuff like that.... as I said, it makes no sense to define the requirements. sometimes, it makes sense to just add something without looking (e.g. it was done by someone trusted), sometimes a real review is needed.
Marcus